A New Methodology Based on q-Entropy for Breast Lesion Classification in 3-D Ultrasound Images

Abstract

Classification of breast lesions is clinically most relevant for breast radiologists and pathologists for early breast cancer detection. This task is not easy due to poor ultrasound resolution and large amount of patient data size. This paper proposes a five step novel and automatic methodology for breast lesion classification in 3-D ultrasound images. The first three steps yield an accurate segmentation of the breast lesions based on the combination of (a) novel non-extensive entropy, (b) morphologic cleaning and (c) accurate region and boundary extraction in level set framework. Segmented lesions then undergo five feature extractions consisting of: area, circularity, protuberance, homogeneity, and acoustic shadow. These breast lesion features are then input to a support vector machine (SVM)-based classifier that classifies the breast lesions between malignant and benign types. SVM utilizes B-spline as a kernel in its framework. Using a data base of 250 breast ultrasound images (100 benign and 150 malignant) and utilizing the cross-validation protocol, we demonstrate system's accuracy, sensitivity, specificity, positive predictive value and negative predictive value as: 95%, 97%, 94%, 92% and 98% respectively in terms of ROC curves and A_z areas, better in performance than the current literature offers.

GeneAI 3.0: powerful, novel, generalized hybrid and ensemble deep learning frameworks for miRNA species classification of stationary patterns from nucleotides

“…In order to extract distinctive and discriminative features, the XC matrices are subjected to normalization, resulting in the transformed matrices XC ′ . Subsequently, the normalized co-occurrence matrix XC ′ is utilized to compute several properties, which include Entropy, Contrast, Energy, Homogeneity, and Dissimilarity 95,[99][100][101] . The mathematical equations for these properties can be found in Table 3.…”

GeneAI 3.0: Powerful, Novel, Generalized Hybrid and Ensemble Deep Learning Frameworks for miRNA Classification of species-specific Stationary Patterns from Nucleotides

“…Subsequently, the normalized co-occurrence matrix 𝑿𝑪 ′ is utilized to compute several properties, which include Entropy, Contrast, Energy, Homogeneity, and Dissimilarity [78][79][80][81]. The mathematical equations for these properties can be found in Table 11.…”

CAD Algorithms for Solid Breast Masses Discrimination: Evaluation of the Accuracy and Interobserver Variability

“…With the development of computer applications, many imaging tools based on computer-aided diagnosis (CAD) technologies were developed to enhance the physician's diagnostic accuracy. Encouraged by the promising results of CAD for mammographic interpretation, many research groups have started to investigate the effect of CAD for the interpretation of US images (Rodrigues et al 2006;Shen et al 2007). Some CAD approaches have been proposed to assist radiologists for breast mass discrimination (Sahiner et al 2007;Drukker et al 2008).…”